Self‐assembled aggregation of peptides and proteins into regular amyloid fibrils is associated with several neurodegenerative diseases. In case of Alzheimer's disease proteolytic cleavage products of the amyloid precursor protein form pathological amyloid‐beta fibrils in a nucleation and propagation phase. The molecular details and thermodynamic driving forces of amyloid formation are not well understood, but are of high relevance for potential pharmacological interference. We used atomistic binding free energy simulations to calculate the free energy of protofilament propagation by an additional Aβ9–40 peptide binding to the protofilament tip. It requires sampling of relevant conformational transitions which is challenging since the monomeric Aβ9–40 peptide is intrinsically disordered. However, the convergence of umbrella simulations can be enhanced by applying additional restraining potentials on the axial, orientational and conformational degrees of freedom. The improved convergence leads to a much closer agreement with experimental binding free energy data compared to unrestrained umbrella sampling. Moreover, the restraining approach results in a separation of contributions to the total binding free energy. The calculated contributions indicate that the free energy change associated with the restriction of conformational freedom upon propagation makes a large opposing contribution of higher magnitude than the total binding free energy. Finally, optimization of the approach leads to further significant reduction of the computational demand which is crucial for systematic studies on mutations, denaturants and inhibitors in the fibril propagation step.

中文翻译：

---

高效准确地计算 Aβ9–40 原丝生长的结合自由能

肽和蛋白质自组装成规则的淀粉样原纤维与多种神经退行性疾病有关。在阿尔茨海默氏病的情况下，淀粉样前体蛋白的蛋白水解产物在成核和增殖阶段形成病理性淀粉样蛋白-β原纤维。淀粉样蛋白形成的分子细节和热力学驱动力尚不清楚，但与潜在的药理干扰高度相关。我们使用原子结合自由能模拟来计算附加 Aβ 的原丝传播自由能9–40肽与原丝尖端结合。它需要对相关构象转变进行采样，这具有挑战性，因为单体 Aβ9–40肽本质上是无序的。然而，通过在轴向、方向和构象自由度上施加额外的约束势，可以增强伞模拟的收敛性。与无约束伞采样相比，改进的收敛性导致与实验结合自由能数据更加一致。此外，限制方法导致对总结合自由能的贡献分离。计算的贡献表明，与传播时构象自由的限制相关的自由能变化产生了比总结合自由能更大的相反贡献。最后，该方法的优化导致计算需求的进一步显着减少，这对于原纤维增殖步骤中突变、变性剂和抑制剂的系统研究至关重要。